Selecting an appropriate interestingness measure to evaluate the correlation between Chinese medicine syndrome elements and symptoms

Abstract

To select the best interestingness measure appropriate for evaluating the correlation between Chinese medicine (CM) syndrome elements and symptoms, 60 objective interestingness measures were selected from different subjects. Firstly, a hypothesis for a good measure was proposed. Based on the hypothesis, an experiment was designed to evaluate the measures. The experiment was based on the clinical record database of past dynasties including 51 186 clinical cases. The selected data set in this study had 44 600 records. Cold and heat were selected as the experimental CM syndrome elements. Three indicators calculated according to the distances between two CM syndrome elements were obtained in the experiment and combined into one indicator. The Z score, ϕ-coefficient, and Kappa were selected from 60 measures after the experiment. The Z score and ϕ-coefficient were selected according to subjective interestingness. Finally, the ϕ-coefficient was selected as the best measure for its low computational complexity. The method introduced in this paper may be used in other similar territories.